多用户无线信能同传系统能效优化方法

针对多用户无线信能同传系统下行链路,研究多用户无线信能同传系统能效优化方法,提出了一种改进型Dinkelbach方法。首先利用Dinkelbach方法将分式问题转换为减式问题,再通过引入冗余变量将和速率最大化问题等价为加权均方误差最小化相关问题,并利用凹凸优化的思想将该问题近似为二阶锥规划凸问题,最后利用块坐标算法对此问题进行迭代求解得出改进型Dinkelbach方法。仿真结果表明,该方法能有效改善系统能量效率。     

基于流形正则化非负矩阵分解的高光谱数据降维

采用基于流形正则化非负矩阵分解(MR-NMF)的高 光谱数据降维方法。新方 法通过构建样本的近邻图描述数据几何结构,然后将其作为正则项加入NMF的目标函 数中进行组合优化。在真实的高光谱数据集HYDICE上进行的实验结果表明,新方法能 够提高高光谱图像分类的精度。     

Convolutional neural network based deep conditional random fields for stereo matching

Stereo matching has been studied for many years and is still a challenge problem. The Markov Random Fields (MRF) model and the Conditional Random Fields (CRF) model based methods have achieved good performance recently. Based on these pioneer works, a deep conditional random fields based stereo matching algorithm is proposed in this paper, which draws a connection between the Convolutional Neural Network (CNN) and CRF. The object knowledge is used as a soft constraint, which can effectively improve the depth estimation accuracy. Moreover, we proposed a CNN potential function that learns the potentials of CRF in a CNN framework. The inference of the CRF model is formulated as a Recurrent Neural Network (RNN). A variety of experiments have been conducted on KITTI and Middlebury benchmark. The results show that the proposed algorithm can produce state-of-the-art results and outperform other MRF-based or CRF-based methods.     

数字基带传输系统的FPGA设计与实现

为了提高系统的集成度和可靠性,降低功耗和成本,增强系统的灵活性,提出一种采用非常高速积体电路的硬件描述语言（VHDL语言）来设计数字基带传输系统的方法。详细阐述数字基带传输系统中信号码型的设计原则,数字基带传输系统中信号编码原理和译码原理;采用硬件描述语言来设计数字基带信号编码器和译码器并进行仿真;采用原理图设计方法设计数字基带传输系统并仿真;整个系统的设计在QuartusⅡ平台上完成,并在Ahera公司的ACEX1K-EP1K30TC144-1芯片上实现。     

静电场中静电屏蔽的Ansoft求解

静电屏蔽是“电磁场与电磁波”课程静电场教学中的一个重要知识点。本文基于Ansoft工程电磁场有限元分析软件,给出金属罩接地与不接地情况下,对空间电场分布的影响。分析结果表明,闭合的金属罩可屏蔽内部空间以防电磁干扰,接地的金属罩能实现全屏蔽。对场的空间描绘可增加学生对场的感性认识,提高教学效果。     

Optimal design of a tunable electromagnetic shunt damper for dynamic vibration absorber

Dynamic vibration absorber (DVA) is an effective device for suppressing resonant vibration of noisy machineries and structures. However, the optimum design of DVA requires precise tuning of the damping force in the DVA, which unfortunately is often not practical and prone to changes of working conditions. In this paper, a tunable electromagnetic shunt damper (EMSD) with different opposing magnet pairs configurations is tested for the optimum design of DVA. The optimum magnet pairs configuration is derived to provide the maximum damping force in the DVA. Both simulations and experiments are conducted to verify the damping coefficient variation with the number of magnet pairs in the EMSD. The experimental optimization procedure of the DVA is designed according to the fixed-points theory. The damping force generated by the EMSD can be readily adjusted by varying the external resistance of the EMSD. This is the first experimental implementation report of the optimization procedure described in the fixed-points theory. The proposed tunable EMSD can conveniently allow for onsite optimal tuning of DVA. The proposed design methodology provides fine tuning of the damping coefficient of EMSD to achieve robust optimal DVA performance, even when subject to changes of external parasitic damping.     

一种开关磁阻电机模糊PI的直接瞬时转矩控制

噪音是开关磁阻电机应用中的一大难题。针对开关磁阻电机运行过程中转矩脉动较大的问题,采用有限元计算方法所得到的电机模型,在MATLAB中进行开关磁阻电机模糊PI直接瞬时转矩控制的仿真实验。采用Ansoft电磁场仿真软件,搭建了三相4/6开关磁阻电机,并将得到的磁链数据导入MATLAB电机模型中。随后,利用该模型搭建模糊PI直接瞬时转矩控制的Simulink仿真实验平台。实验结果发现,系统在0.01 s时开始响应,转矩浮动在0.05~0.1 Nom之间。该结果表明,将有限元得到的电机模型应用到模糊PI直接瞬时转矩控制的仿真实验中时,可以提高系统响应,并有效抑制转矩脉动。     

基于曲折网格表面的二维通信技术

二维通信系统(Two-dimensional Communication System, 2DCS)是一种以物体表面为通信介质的新型 通信技术,可以提供无线供电以及安全、高速的数据传输服务。为了改善二维通信中的供电性能,文中提出了一种 增强功率传输性能的二维通信导波板,该导波板包含三层:导电层、介质层和具有曲折网格的表面层。微波通过同 轴导波耦合器馈入二维通信导波板,可以在呈电感性的曲折网格表面任何一点耦合出表面层。经电磁仿真和实际 测试验证,与以前的二维通信导波板相比,曲折网格结构和阻抗匹配的同轴导波馈电器改善了二维通信导波板的传 输性能和电磁场分布的空间均匀性。     

Searching proper oligothiophene segment as centre donor moiety for isoindigo-based small molecular photovoltaic materials

Systematic studies on a family of photovoltaic molecules are important for fundamentally understanding the basic principles and the key structural factors that govern their photovoltaic performance. In this work, a series of D²-A-D¹-A-D² type small molecules with isoindigo as acceptor (A), oligothiophene as donor (D¹), and 5-hexyl thiophene as D² were designed, synthesized and studied. The number of thiophene unit in oligothiophene segment is systematically varied from 0 to 4. It has been found this structural parameter have significant influence on their phase transition behaviours, light absorption properties, frontier orbital energy levels, molecular packing structures in both neat and blending films with [6,6]-phenyl-C₆₁-butyric acid methyl ester (PC₆₁BM), charge mobility, phase separation morphology, and as well as photovoltaic performances. The compounds bearing odd number of thiophene unit displayed much better photovoltaic performance than those having even number, while the best performance was given by that having terthiophene as D¹ unit.     

Efficient distributed traffic offloading algorithm based on SDN architecture

For addressing the problem of mobile data traffic offloading,network service cost minimization problem model was proposed,which considered the cost of base stations and the consumption of access points under the SDN architecture.Due to the defects of convergence speed and privacy security in the traffic offloading algorithm based on dual decomposition,the traffic offloading algorithm based on proximal Jacobian ADMM was proposed and an implementation scheme to ensure the privacy and security was designed.Meanwhile,the simulation results show that the algorithm based on the proximal Jacobian ADMM is superior to the dual decomposition algorithm in terms of convergence speed.